Forecast Customer Service for Q4 and Beyond

Jon Tucker, CEO at helpflow.com
October 12, 2020
Forecasting customer service volume for Q4 has always been challenging. But e-commerce growth due to Covid, along with logistics nightmares that are happening, makes forecasting even harder.

In this post by Jon Tucker from HelpFlow, you'll learn how to easily and accurately forecast customer service volume in any environment. HelpFlow provides 24/7 live chat and customer service teams to over 100 stores and will share the insights from their experience here

For many industries in e-commerce, Covid has brought a huge amount of growth for 2020 compared to last year. According to the US Census Bureau, overall e-commerce grew 31.85% from Q1 to Q2 of 2020  due to a big chunk of brick and mortar retail sales moving online. According to a report from Adobe Analytics, ecommerce spending in May alone surged 78% year over year. 

While certain industries like food supplies and medical grew a ton directly due to the covid pandemic, there were other industries that also became big winners. According to a Wix.com eCommerce report looking at year over year sales for April and May, Home decor was up 300%+, games & leisure were up aroud 200%, music was up around 125%, etc.  

But supply chain issues with product and shipping issues with USPS and other carriers have made management harder. 

  • USPS shipping time delays have increased in some cases by 42.03% according to GoShippo
  • Other carriers like UPS and FedEx have had the same issues with on-time delivery dropping from 90% or higher to 86% and 81%, respectively. 

The delays became more and more frequent as Covid shutdowns spread. DigitalCommerce360 visualized the data as shown here. 

Part of the cause of these delays are the sheer volume of additional shipments going out. But carriers also have to comply with new regulations to keep their workers and customers, which brings an added cost. While increased costs for Q4 are common for some carriers on certain delivery options, USPS has held off on these rate increases in past years. That all ended recently with the announcement of USPS rate increases for Q4 shipping, ranging from 5.70% to 13.26%. 

In short, Covid has brought growth to many in e-commerce- but also a lot more difficulties in managing the business. Add on to that the typical craziness of Q4 multiplied due to Covid, and it can become incredibly intimidating to “get ready.” 

Forecasting customer service workload does not need to be complicated, even with the unpredictable nature of Covid. You can forecast customer service volume as a percentage of revenue growth, which is much more predictable than trying to simply “guesstimate” how much workload your team can handle or if you need to hire more people for the Q4 rush.

At HelpFlow, we run 24/7 live chat and customer service teams for over 100 stores. We’ve been through five holiday seasons and have built an incredibly robust forecasting and KPI process. In this post, I will walk through how to simplify Q4 customer service forecasting to make sure you’re ready to rock (and not get rocked) this holiday season. 

But 1st - How Big is Q4 Going to Be for Sales with Covid?

As the pandemic took over in March, every business was scary to be in. Sales slowed down, hiring slowed, and people buckled down for the end of the world. But in late March and early April, things started changing a lot in e-commerce.

Here’s a pretty typical revenue trend that we saw for a wide range of e-commerce clients as the pandemic played out. Notice the massive increase for March to April, continued growth for May, and then a leveling out in June and July to far above pre covid levels. 


Some industries did better than others, and some that were directly related to pandemic-affected niches declined.  But in general, e-commerce grew quite a bit. 

  • For clients that were well-positioned with inventory and a solid advertising engine, many were able to 15x or 20x their year over year growth rate for April to June compared to their year over year growth rate pre-pandemic in February. They didn’t get lucky, they scaled advertising heavily and executed well. 
  • But even for brands that did not rely heavily on paid traffic for growth, most were able to 3 to 5X their year-over-year growth rate for April to June compared to pre-pandemic year over year growth.

Sales Increased with Covid - But So Did Ticket Volume

With this growth, there was also a massive increase in customer service ticket volume. Below are some trends from the Gorgias customer base data set. 

  • From the week of March 16th to the week of April 17th, there was about a 2x increase in ticket volume. 
  • This is similar to the trend of revenue growth shown above - but ticket volume didn’t really come back down over time as sales did. 



How Should Covid Be Factored in to Forecasting

April and May’s growth rates for sales were massive, but things normalized a bit for June and July. If you’re trying to gauge how much bigger Q4 will be because of Covid compared to last year, June and July Covid growth is a good place to start. Do some analysis to see how much higher June and July are this year compared to last year, and factor that growth rate into your Q4 planning.

Based on the average pandemic fueled growth rate we saw across a wide range of clients, planning for 3x to 5x what you would have normally forecasted for Q4 2020 is a reasonable rule of thumb.

But before you get too excited with the financials in your spreadsheet, remember that you’re going to need to deal with all the management and customer service volume that’s going to come with this growth. Plus, with all the logistics nightmares and shipping carrier issues that have popped up this year, the workload could also be much higher than it would’ve been without Covid.

A lot of brands got surprised by the volume increase during Covid. Everyone knows Q4 is going to be big, but you don’t have to be surprised with “how big?”. 

  • Predicting the future workload ahead of time is much better than colliding with it and having to throttle your growth. 
  • The forecasting process we will walk through now will help you see how Q4 will play out for your business so you can be ready.

So let’s get into it and forecast the future...

The Simple Secret to Forecasting Customer Service Volume

Determining whether you need more customer service agents should not be based on merely reacting to your team asking for help. It should be based on data, like every other part of scaling an e-commerce business.

The challenging part here is figuring out which data in your business can be used to forecast how many agents you need. We’ve simplified this process for clients we run customer service for by using just two inputs: sales transaction volume and ticket volume for a period of time. 

This can be used to calculate your “Transaction to Ticket Ratio.” In short, you can forecast the number of tickets created for every 100 transactions and be extremely confident that this will hold steady as growth in transactions happens. Yes, there are ways to make your customer service process more efficient to lower this ratio - but knowing where it stands now will enable you to forecast ticket volume based on your sales forecast. 

Step #1 - Calculate your Transaction to Ticket Ratio

Calculating your transaction to ticket ratio is simple. All you need is transaction count for a period of time and ticket volume for the same period of time.

To get your transaction count, simply run a quick report in Shopify. The process will be similar in other e-commerce platforms. 

  1. Click on Reports under the Analytics category and select Sales over time.
  1. Highlight the target date range and select the desired group. 
  1. This will generate the data that includes the number of transactions within the specified time frame.

You can also get this data directly from Google analytics, by using the e-commerce section and seeing transaction volume. This might not be accurate if a significant portion of your transactions are subscriptions that are not tracked in analytics. Still, it will be close enough for now as long as you continue to use this transaction count in future calculations.

How to get transaction count in Google Analytics:

  1. Click on Conversions. Select E-commerce, and then Overview.

  1. Select the target date range. The transaction count will be generated under the Transactions field.

The next data point you need is ticket volume for the same time period. In Gorgias, you can access this easily by following these steps:

  1. Click Overview under the Statistics menu and select the target date range. Click Apply.

  1. This will generate the data on ticket volume for the specified timeframe.

Once you have transactions and ticket volume for the same time period, simply divide tickets by transactions to calculate your ratio. For example, if you had 1000 transactions for the time and 400 tickets, then your ticket ratio is 40. For every 100 transactions, 40 tickets are created. 

What’s Healthy for a Ticket Ratio?

Your specific ticket ratio will depend on many factors, including the sophistication of your customer service process/automation, how effective your fulfillment and delivery process is, etc. 

  • We typically see the ticket ratio anywhere from 30% to 50% for stores that haven’t focused on streamlining customer service. 
  • For stores that have streamlined or use more automation in their HelpDesk workflow, the ticket ratio normalizes to about 20%. For example, across high revenue stores in the Gorgias customer data set, there is a 20% ticket ratio. (This is one of the great benefits that you can get from Gorgias’ sophisticated platform.) 

If you haven’t focused on streamlining customer service and have just been keeping up with growth over months or years at a time, then you’ll probably be surprised at how many orders turn into tickets. It’s definitely something you can bring down overtime with operational improvements in your customer service department, but just benchmarking it for now is important to forecast accurately.

Step #2 Use The Ratio To Project Ticket Volume

Once you have the transaction to ticket ratio, you can forecast customer service ticket volume in a few ways. You can use website traffic, media spend, or revenue projections as the input. Basically, anything that enables you to get a transaction count as an input can be converted into customer service ticket volume using the ticket ratio. 

Here are a few examples:

Let’s say your business is highly driven by paid traffic. If you have a budget set for increased media spend in Q4, you can gauge how many orders this will produce using your cost per acquisition records. For example, if your CPA is $20 and you’re projecting a media spend of $200,000 in November and December, this should produce around 10,000 orders. 

Using a different approach, let’s say you are projecting revenue of $600,000 total for November and December and your average order value is $60. This means you’ll have 10,000 orders. 

Once you have the projected transaction volume, simply use the ratio to convert that into ticket volume. 

For example, if your ticket ratio is 40%, then the 10,000 orders are going to turn into 4000 tickets for your team to handle during November and December.

If you can bring your ticket ratio down to 20%, this cuts the ticket volume and only 2000 tickets. You can see how improving the ticket ratio can have a big impact on the efficiency of your customer service team.

Once you’ve got the forecasted ticket volume, you still need to figure out how many team members are going to be needed to handle the volume. 

Step #3 How Many Agents Are Needed?

There’s nothing more frustrating as a business owner than having to slow down your sales engine because you can’t keep up with orders or customer service volume. At some point, you can’t just throw more people at the problem. People are a finite resource and it takes time to hire and train competent agents. Don’t let yourself end up in that position this season.

As part of running an effective customer service operation, you should have some benchmark set on agents’ realistic capacity to handle tier 1 and tier 2 tickets. With this data, you can convert the forecasted ticket volume we calculated above into a forecasted agent headcount to handle that volume.

Let’s assume you don’t have a complete agent capacity benchmark set yet.

To calculate agent capacity per full-time agent, run a report on the total number of tickets resolved per agent over a specific time period. Be sure to focus your analysis on agents that are processing tickets nearly full time, not part-timers that jump in to help.

Simply click on the Agents option under the Statistics menu.

For each agent, convert this into the number of tickets handled per working day. The specific number will be different for each agent, but running this process will help you see the trends of how many tickets per day a typical agent should be able to handle.

Again, this number will vary a ton depending on your business, your agents, and your customer service operation. 

  • We typically see anywhere from 40 to 60 tickets per day per agent as a healthy benchmark.  
  • Across the Gorgias customer dataset, we see an average of 60 tickets per day per agent, again showing the benefits of a sophisticated HelpDesk platform and workflow. 
  • At peak Covid times, some Gorgias customers were able to handle 100+ tickets per day, though this is something that is tough to sustain over time. 

Once you know your tickets per day per agent benchmark, you can convert the forecasted ticket volume above into a forecasted agent headcount needed for November and December.

Let’s say you are projecting 4000 tickets for your team to handle during November and December and you have an agent capacity of 40 tickets per day. You’re going to need 3 agents staffed during November and December to handle this volume. 

40 tickets per agent per day x 5 days per week x 4.3 weeks per month = 860 tickets monthly capacity per agent. 4000 tickets / 2 months  / 860 tickets per agent per month = 2.33 agents. Round up to 3 so you have buffer for scenarios that may drive ticket volume up (e.g., delayed shipping, running out of stock, promotions and sales)

This is a little simplistic since it assumes the tickets come fairly uniformly during that time, but the basic forecasting process here is sound. 

Customer Support Staffing Calculator

Forecast different scenarios and anticipate the customer support needed to deliver an outstanding shopping experience.

Customer Support Staffing Calculator

Forecast different scenarios and anticipate the customer support needed to deliver an outstanding shopping experience.

at
Customer Support Staffing Calculator

Forecast different scenarios and anticipate the customer support needed to deliver an outstanding shopping experience.

Thank you! Your submission has been received!
Oops! Something went wrong while submitting the form.
Customer Support Staffing Calculator
Customer Support Staffing Calculator

Forecast different scenarios and anticipate the customer support needed to deliver an outstanding shopping experience.

Thank you! A copy of the ebook has been sent in your email inbox. In the meantime, you can access  to the ressource
here
.
Oops! Something went wrong while submitting the form.
Customer Support Staffing Calculator
Customer Support Staffing Calculator

Forecast different scenarios and anticipate the customer support needed to deliver an outstanding shopping experience.

Thank you!  You can access  to the ressource
here
.
Oops! Something went wrong while submitting the form.

Step #4 - Don’t Underestimate the Covid Factor

I mentioned at the start of this post how Covid has helped many e-commerce businesses have their best year ever. That’s great, but its Covid.  It has wreaked havoc on the supply chain logistics for many of these businesses and caused massive headaches in customer service due to delivery issues and delays.

The best way to accommodate for this in your forecast is to use ticket volume and ticket ratios from recent time periods, especially where logistics has been a problem. You might even consider Inflating your ticket ratio a little bit to accommodate for more logistics issues happening this holiday season.

Don’t assume logistics is going to be easier in the fourth quarter than it has been all year. Work hard to improve where you can, but be ready for things to get worse before they get better. By factoring us into your forecast as a variable, you’ll be ahead of the game.

Need Help Maximizing Q4? 

The way you handle customer service in Q4 has a massive impact on sales. 

  • Based on analyzing Q4 2019 Gorgias data across all customers, a sub 10 minute response time on tickets increases conversion rate around 10%. 
  • Answering live chat questions in less than 2 minutes increased conversion rate by as much as 50%!
  • At HelpFlow, we run live chat teams for clients and have seen a massive dropoff in chat conversion rate when your first response time is over 10 seconds. 

If your team is overloaded with customer service volume, responding to tickets in < 10 minutes is going to be near impossible and answering live chats in seconds is definitely out of the question. 

Here’s what to do next:

  1. At HelpFlow, we provide 24/7 live chat and customer service teams to over 100 e-commerce stores (i.e. our agents answer visitor questions on chat and handle email tickets, 24/7). We’ve gone through 5 holiday seasons with thousands of tickets per day and millions of dollars at stake. If you need help nailing Q4, take look at our customer service teams here.
  2. Already have a solid team and want to bulk up your customer service infrastructure? If you’re not already using Gorgias, then that’s the first step. Signup for a free trial here.

Have a great Q4!

Want to Deliver an Outstanding Support Experience?

Try Gorgias! 7-day free trial, no credit card required & instant set-up.

Join the 

4500

+ ecommerce
brands that use Gorgias
every day

Thank you! Your submission has been received!
Oops! Something went wrong while submitting the form.
Get a live demo